I am a Member of Technical Staff at the Centre for AI Research and Excellence (CAIRE) at Statistics Canada. I build full-stack AI systems and the evals that make them safe to deploy in confidential-data settings.
My current focus is AI security, applied mechanistic interpretability, and uncertainty quantification for AI systems.
- Won International Association for Official Statistics Young Statisticians Prize 2025 for a Bayesian framework integrating LLMs with uncertainty quantification.
- I contribute to guidance and governance work with UNECE, G7 GovAI, and the International Statistical Institute.
- Fine-tune Auditor (SAE affordances): adds SAE-based model diffing to fine-tuning auditing agents, exposed via an MCP tool server for end-to-end agentic audits.
- Probability Lab: interactive probability and statistics learning platform (code, course context).
- Reinforcement Learning Labs: a 6-part notebook series (notebooks, master thesis).